Due　to　the　high　flexibility　and　high　maneuverability,　UAVs　play　an　important　role　in　constructing　the　space-terrestrial　integrated　network　(STIN).　However,　in　STIN-assisted　sparse　IoT　scenarios,　a　single　UAV　cannot　provide　seamless　and　long-term　services　due　to　its　limit　cruising　duration,　hence　the　cooperation　of　multiple　UAVs　becomes　a　remedy.　This　paper　studies　the　cooperative　strategy　of　multiple　rechargeable　UAVs　under　the　constraints　of　supporting　seamless　and　long-term　information　services　for　ground　IoT　nodes.　Since　the　proposed　problem　is　a　mixed-integer　non-convex　problem,　a　sequential　convex　optimization　technique　based　iterative　algorithm　is　proposed　to　jointly　optimize　the　IoT　node　assignment　and　UAV　trajectory.　Simulation　results　show　the　effectiveness　of　our　proposed　method.　©　2019　IEEE.